Biomarker development in the precision medicine era: lung cancer as a case study

Ashley J. Vargas,Curtis C. Harris
DOI: https://doi.org/10.1038/nrc.2016.56
IF: 78.5
2016-07-08
Nature Reviews Cancer
Abstract:Key PointsPrecision medicine seeks to identify and classify individual patients so that optimal treatment decisions can be made. Nations are recognizing this area of research by developing national cohorts from which to collect data and developing regulatory guidelines on biomarkers.The cost of genetic sequencing and other 'omics' technologies has decreased while the quality of data they generate has increased. Thus, immense amounts of molecular data are being derived from cohort studies to begin developing new biomarkers to classify patients into subtypes.The development of biomarkers is largely limited by the following factors: low statistical power in rare subtypes; risk of false-positive findings in studies that do not validate their findings in a separate cohort and/or conduct concomitant mechanistic experiments; and technical reproducibility concerns.Genomics and protein immunohistochemistry have led the way for developing biomarkers. Other molecular measurements (for example, metabolomics and microbiomics) are still in preliminary stages and are often not validated in another cohort.Integrating different types of molecules into a biomarker panel, along with other patient data, is the future of precision medicine; however, the sheer number of potential combinations of data types complicates the concerns about statistical power and reproducibility.Currently, improved biomarkers are needed to differentiate lung nodules identified by new US national screening recommendations into non-cancer, cancer with poor survival probability and cancer with higher survival probability subtypes, to provide thousands of individuals with precise treatment decisions.
oncology
What problem does this paper attempt to address?